Tannoy System 600 Speaker Review

[MZKM]

Can you calculate the Preference rating of the EQ-ed version if I send you the filter?

What does that data look like?

 

[QMuse]

What does that data look like?

I can supply corrected on-axis, LW, ER and SP exported from REW. Here is how it looks for on-axis data.

 

[MZKM]

Repeating what I already know is not helpful. What is helpful is showing how the extra effort is worth it at the expense of measuring another speaker. And remember, it is not just time to measure but capture the results, format, annotate and publish. And field questions about it like I am doing now.

If you ever decide to, just plug the rough sensitivity @2.83V into this pre-ready Wolfram function I just put together. It’s set to achieve 100dB, you can of course change it. Or, since you are measuring at 1/3m does it need to be altered (~9.5dB difference)? Also, how are you measuring a 1/3m distance, just using 13 inches?

 

[MZKM]

I can supply corrected on-axis, LW, ER and SP exported from REW. Here is how it looks for on-axis data.

I’m not at home for the weekend, so doing this all on my phone. If you could use the same CEA-2034 and Estimated In-Room files in the original zip file but just with the different data, then I can easily import it (So, load the originals in Sheets/Excel, replace with the EQ’d data, export as CSV).

EDIT: If you don’t have the corrected in-room, I’d have to calculate that, which unless I can find where I manually calculated it before, it would have to wait till I get home.

 

[MZKM]

Power = voltage * voltage /impedance. The impedance varies here but using 5 ohm, the power would be 100/5 = 20 watts.

SPL increase would be 12 dB.

Isn’t it +11dB?

20•log10(10/2.83)≈11dB

Conversely:
2.83^2 /5 ≈ 1.6W
10^2 /5 = 20W
20/1.6 = 12.5 ratio
10•log10(12.5)≈11dB

 

[QMuse]

I’m not at home for the weekend, so doing this all on my phone. If you could use the same CEA-2034 and Estimated In-Room files in the original zip file but just with the different data, then I can easily import it (So, load the originals in Sheets/Excel, replace with the EQ’d data, export as CSV).

EDIT: If you don’t have the corrected in-room, I’d have to calculate that, which unless I can find where I manually calculated it before, it would have to wait till I get home.

Ok, I'll give it a try.

I have corrected in-room data.

 

[thewas]

Exceptionaly smooth directivity indexes allow for effective EQ. This filter variant has been optimised for flat on-axis response:

On-axis:

View attachment 53310

LW & PIR:

View attachment 53311

If I would own these loudspeakers I wouldn't EQ the almost 10dB dip though as it would take the increased distortions to an even higher point, but the rest can be nicely EQed as you show. The good thing is that our hearing is also not very sensitive to narrow dips, opposite to narrow peaks.

 

[DeeJay]

I don't agree. Current measurements are done according to CEA-2034 standard which is based on parameters proven to be the most relevant for SQ.

Dr. Floyd Toole: "A spinorama shows linear performance. Collectively the curves reveal evidence of resonances, bandwidth, spectral balance, smoothness and directivity.​

...​

Then there are the other variables: power handling, power compression and non-linear distortions. These are not shown in the spinorama, but competent design engineers measure them in the selection and design process. Tweeters with identical domes, for example, can exhibit very different behaviors because of differing motor designs - which are invisible.​

Summing up, unless there is serious misbehavior in one of these other variables, the linear behavior described by a spinorama is the dominant clue about potential sound quality."​

If you have two speakers with identical spinorama and power handling, how would you know which one has better SQ playing at concert levels? How do you know if there is no "serious misbehaviour"? I don't think you can without additional measurements.

 

[maty]

Tannoy System 800 never sounded right
https://www.diyaudio.com/forums/multi-way/265772-tannoy-system-800-sounded.html

-> https://www.diyaudio.com/forums/multi-way/265772-tannoy-system-800-sounded-post4141476.html

I did some in-room measurements recently while trying to figure out the problem, that showed a mysterious notch in the SPL at about 1.5kHz. Now you mention it, a HF driver wired backwards could well explain that.

-> https://www.diyaudio.com/forums/multi-way/265772-tannoy-system-800-sounded-post4141772.html

I reversed the wires to the tweeter and repeated the measurement. The notch disappeared and on listening to music, the midrange output seemed a lot better.

 

[maty]

-> https://www.diyaudio.com/forums/multi-way/265772-tannoy-system-800-sounded-post5675683.html

 

[DeeJay]

As it relates to listener preference, yes. And that, based on every research shows that THD is not a factor.

In listening tests, I comment on power handling.


Quality is superbly there in the spinorama measurements. They are state of the art and comply with the best of what we know to characterize listener preference. Your request is to pile on more quantity with little value relative to work already being done.

It is not like we are sitting on data from 1000 speakers already to spend time drilling down on a few. We need to get good coverage of performance from all the major brands to know who is and who is not building good speakers.

Are you after fidelity or listener preferences? Are you saying that THD of speakers are not important but amplifiers are?
Listener preferences are different. Absolute fidelity should be the goal.

Kevin Voecks: "One thing that everyone can take to the bank is that poor Spinorama measurements guarantee the speaker will not sound good."

Yes, but does not guarantee that it will. Also, see Dr. Floyd Tolle comment in my previous post. Sorry, but a comment on power handling is not enough.

 

[maty]

You should ask the owner if you can open it and reverse the tweeter.

BTW, thin thick walls.

https://www.manualslib.com/manual/481948/Tannoy-System-600.html?page=9#manual

36 mm MDF and 19 mm HDF!

 

[maty]

https://www.manualslib.com/manual/481948/Tannoy-System-600.html?page=8#manual

 

[q3cpma]

First time I thought "something is probably broken in here".

Dr. Floyd Toole: "A spinorama shows linear performance. Collectively the curves reveal evidence of resonances, bandwidth, spectral balance, smoothness and directivity.​

...​

Then there are the other variables: power handling, power compression and non-linear distortions. These are not shown in the spinorama, but competent design engineers measure them in the selection and design process. Tweeters with identical domes, for example, can exhibit very different behaviors because of differing motor designs - which are invisible.​

Summing up, unless there is serious misbehavior in one of these other variables, the linear behavior described by a spinorama is the dominant clue about potential sound quality."​

If you have two speakers with identical spinorama and power handling, how would you know which one has better SQ playing at concert levels? How do you know if there is no "serious misbehaviour"? I don't think you can without additional measurements.

Exactly that. THD is not that important unless we're talking about high amount in mid or high frequencies; can also be useful to choose between speakers already having good spins, I guess. It's also useful since it's mostly what signals that the speaker is exceeding its physical capabilities.

 

[LTig]

As it relates to listener preference, yes. And that, based on every research shows that THD is not a factor.

In listening tests, I comment on power handling.


Quality is superbly there in the spinorama measurements. They are state of the art and comply with the best of what we know to characterize listener preference. Your request is to pile on more quantity with little value relative to work already being done.

It is not like we are sitting on data from 1000 speakers already to spend time drilling down on a few.

This is correct, but I also agree that a SOTA speaker also needs to have proper SPL capabilities.

So as a compromise maybe you could add SPL- based distortion tests only for speakers with very good spinoramas. Or for speakers especially specced as being able to play very loud. You would spare a lot of time as long as there are lots of bad speakers around, but still supply a full set of measurements where it really matters.

 

[QMuse]

Dr. Floyd Toole: "A spinorama shows linear performance. Collectively the curves reveal evidence of resonances, bandwidth, spectral balance, smoothness and directivity.​

...​

Then there are the other variables: power handling, power compression and non-linear distortions. These are not shown in the spinorama, but competent design engineers measure them in the selection and design process. Tweeters with identical domes, for example, can exhibit very different behaviors because of differing motor designs - which are invisible.​

Summing up, unless there is serious misbehavior in one of these other variables, the linear behavior described by a spinorama is the dominant clue about potential sound quality."​

If you have two speakers with identical spinorama and power handling, how would you know which one has better SQ playing at concert levels? How do you know if there is no "serious misbehaviour"? I don't think you can without additional measurements.

No, you cannot. But considering limited human resources it may be of greater practical value to get more spinorama of various speakers than to get more detailed measurements of smaller number of speakers.

 

[tomtoo]

Something is wrong with this speaker. I can't believe Tannoy makes there own FR diagramms from pure Phantasie.

 

[JIW]

Isn’t it +11dB?

20•log10(10/2.83)≈11dB

Conversely:
2.83^2 /5 ≈ 1.6W
10^2 /5 = 20W
20/1.6 = 12.5 ratio
10•log10(12.5)≈11dB

Yes, using the voltage ratio is sufficient and gives the 11 dB you indicate.

However, your second calculation is identical to the first. Since the impedance is unchanged, the only difference is the voltage and the ratio of the powers is the ratio of the squared voltages. Further, in the level calculation based on power the argument (the power ratio) is raised to half the power (i.e. 1) than the argument is raised in the level calculation based on voltage (i.e. 2) (the coefficient before the logarithm for the power based calculation is half the coefficient for the voltage based calculation). Thus, the two different ways of calculation merely move the power to which the voltage ratio is raised (2) to a different place in the equation.

Generally for x>0 and 0<b, b≠1:
a•ln(x) = ln(x^a)
log_b(x) = ln(x)/ln(b).
Further,
c•d•ln(x) = d•ln(x^c).

Thus,
10•log_10(x^2) = 2•10•log_10(x) = 20•log_10(x).

In this case, b = 10, d = 10/ln(10) and c is the power to which the voltage ratio (x) is raised, i.e. c = 2.

Specifically:
20•log_10(10/2.83) = 10•log_10((10/2.83)^2) = 10•log_10(10^2/2.83^2) = 10•log_10(100/8) = 10•log_10(12.5) (recall that 2.83 is 8^0.5 or sqrt(8)).

 

[QMuse]

If I would own these loudspeakers I wouldn't EQ the almost 10dB dip though as it would take the increased distortions to an even higher point, but the rest can be nicely EQed as you show. The good thing is that our hearing is also not very sensitive to narrow dips, opposite to narrow peaks.

That dip at 1500Hz is quite wide and it is very near to the area where our ears are most sensitive. That dip is also preceded with wide peakso those 2 should be considered together. Keep in mind that only 2 pages of Toole's 400 pages book are about non-linear distorion, and for a good reason - in listening tests it turned out masking takes care of most of it, which is not the case with linear distortion. And that is why scoring penalises peaks and dips but doesn't take into consideration non-linear dostorion. For that reason I choosed to correct it.

 

[QMuse]

I’m not at home for the weekend, so doing this all on my phone. If you could use the same CEA-2034 and Estimated In-Room files in the original zip file but just with the different data, then I can easily import it (So, load the originals in Sheets/Excel, replace with the EQ’d data, export as CSV).

EDIT: If you don’t have the corrected in-room, I’d have to calculate that, which unless I can find where I manually calculated it before, it would have to wait till I get home.

Here are the files containing corrected data, hopefully I did it right so you'll be able to load them.